Abstract: This is an application for a K22 award for Dr. John Paul Shen, a medical oncologist currently at the University of California, San Diego. Dr. Shen is establishing himself as a young translational investigator in the field of cancer genomics. This K22 award will provide Dr. Shen with the resources and training to accomplish the following objectives; (1) implement advanced computational methods on genome scale datasets (2) become an expert in functional genomics, (3) achieve proficiency experimenting in mouse models of cancer, (4) successfully manage an independent laboratory. To achieve these objectives, after accepting a faculty position Dr. Shen will assemble a diverse advisory committee including experts in bioinformatics, experimental cancer biology, and clinical oncology. It was proposed by many that the ability to sequence a tumor genome, now made possible by next- generation sequencing, would bring about a new era of precision oncology in which chemotherapy choices would be individualized to match a single tumor and patient. However, the use genomic information in clinical practice remains limited by the fact that currently very few mutations are associated with response to a specific drug. This is particularly true in Gastrointestinal (GI) malignancies, where there are few targeted therapy options and few effective biomarkers help guide chemotherapy selection. Dr. Shen seeks to address this pressing need by employing high-throughput functional genomic methods to identify tumor specific vulnerabilities that could be exploited therapeutically. Recognizing that there will be great heterogeneity from one tumor to the next, even within the same cancer type, the functional genomic data created here will be combined with systems biology methods to identify how the vulnerabilities of each unique tumor can be predicted with information readily available to a clinical oncologist. Using network-based machine learning methods applied to chemo-genomic viability data in molecularly characterize cell lines it is expected that predictive biomarkers will be identified for both novel targeted agents and currently used chemotherapy drugs. This will allow oncologists to design individualized chemotherapy regimens for each patient.